Sheldon Wang
Staff Engineer · Agentic Systems & AI Infra
- US
- Meta, ex-Snap
- Github
- Stackoverflow
- YouTube
You May Also Enjoy
Max Effort Is a Bad Default: Start Low, Escalate Only on Verified Failure
8 minute read
Published:
Max reasoning effort can cost 42x more on tasks every effort level solves — and still lose tasks low effort solved. Start low, escalate only on verified failure.
最高推理强度是个坏默认值：先低强度跑，验证失败再升级
1 minute read
Published:
在所有强度都能解出的任务上，最高推理强度最高贵 42 倍，还会丢掉低强度已经解出的题。先低强度跑，验证失败再升级。
One Valid Config Change, 85% of the Network Down: Fastly’s June 8, 2021 Outage and the Bug That Slept for 27 Days
24 minute read
Published:
One customer pushed a valid configuration change. A bug deployed 27 days earlier woke up, and 85% of Fastly’s network started returning errors. Detection took one minute; finding the one guilty configuration took forty — because a global symptom carries no clue about which input caused it.
一次合法的配置变更，85% 的网络瘫痪：Fastly 2021 年 6 月 8 日宕机与那只沉睡了 27 天的 Bug
3 minute read
Published:
一位客户推送了一次合法的配置变更。一个 27 天前部署的 Bug 被唤醒，Fastly 全网 85% 开始返回错误。发现只用一分钟；找到那一个有问题的配置却用了四十分钟——因为一个全球性的症状不会告诉你，是哪一个输入造成的。